gplab-users

[gplab-users] User manual

[Ali D. M. <dan...@ya...>]

Dear all,
Happy new year 2021

I have started to use GPLAB 4.04.
Is there any user manual for this new version?
I found only Version 3 (2007) online.
Thanks a lot for your reply in advance.

Ali


Sent from Mail for Windows 10

Re: [gplab-users] Confusion in using EVALIN()

[Sara S. <sa...@fc...>]

It depends... If you're evaluating a candidate solution on the training 
set, use the training variables, otherwise use the test variables. I'm 
curious regarding the usage of evalin instead of eval, is it faster?

Thanks,
Sara


On 06-09-2018 12:09, Atif wrote:
> Hi All,
>            When we use */evalin()/* instead of *eval() *in our fitness 
> function, then definitely we need to provide the variables (X1, X2, 
> ..) from the dataset file in the workspace. As I've two files 
> (training and testing) so which file I should use for creating these 
> variables? Training, Testing or a combination of both?
>
> Thank you in anticipation !!!
>
> -- 
> /Atif/
> /Research Assistant/
> /Department of Computer Science/
> /Namal College, Pakistan /
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users

[gplab-users] Confusion in using EVALIN()

[Atif <ati...@na...>]

Hi All,
             When we use *evalin()* instead of *eval() *in our fitness
function, then definitely we need to provide the variables (X1, X2, ..)
from the dataset file in the workspace. As I've two files (training and
testing) so which file I should use for creating these variables? Training,
Testing or a combination of both?

Thank you in anticipation !!!

-- 
*Atif*
*Research Assistant*
*Department of Computer Science*
*Namal College, Pakistan *

Re: [gplab-users] Confusion in Fitness function in GPLab

[Atif <ati...@na...>]

Thanks Ma'am Sara,
          In my experiment, *best so far* and *test fitness* are always
coming the same. Why is it so?

On a separate note, I'm using Daily Stock Data. Accordingly, I don't have
output files. As GPLab requires both input and output files for the
successful run so I've given the same input file as my output file, (see
below code lines)

     p.datafilex='training.txt';
     p.datafiley='training.txt';
     p.usetestdata=1;
     p.testdatafilex='testing.txt';
     p.testdatafiley='testing.txt';

However, I don't use these output files in the fitness function.

Thanks Indeed !!!


On 19 August 2018 at 22:27, Sara Silva <sa...@fc...> wrote:

> ind.fitness stores the fitness.
>
> ind.result stores the predicted values for all the samples (all the rows
> in your data input file). The fitness is calculated based on these (RMSE,
> or other).
>
> Cheers,
> Sara
>
>
>
> On 15-Aug-18 5:19 AM, Atif wrote:
>
> Dear All,
>                  I hope everyone is doing fine. I'm confuse which
> attribute defines the fitness value.
>    *ind.fitness   OR   ind.result ?*
>
> I couldn't find the differences in manual.
> Thank you in anticipation !!!
>
>
>
>
> --
> *Atif*
> *Research Assistant*
> *Department of Computer Science*
> *Namal College, Pakistan *
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
>
> _______________________________________________
> gplab-users mailing lis...@li...https://lists.sourceforge.net/lists/listinfo/gplab-users
>
>
>
> ------------------------------------------------------------
> ------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users
>
>


-- 
*Atif*
*Research Assistant*
*Department of Computer Science*
*Namal College, Pakistan *

Re: [gplab-users] Confusion in Fitness function in GPLab

[Sara S. <sa...@fc...>]

ind.fitness stores the fitness.

ind.result stores the predicted values for all the samples (all the rows 
in your data input file). The fitness is calculated based on these 
(RMSE, or other).

Cheers,
Sara



On 15-Aug-18 5:19 AM, Atif wrote:
> Dear All,
>                I hope everyone is doing fine. I'm confuse which 
> attribute defines the fitness value.
> */ind.fitness/  OR /ind.result/ ?*
>
> I couldn't find the differences in manual.
> Thank you in anticipation !!!
>
>
>
>
> -- 
> /Atif/
> /Research Assistant/
> /Department of Computer Science/
> /Namal College, Pakistan /
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users

[gplab-users] Confusion in Fitness function in GPLab

[Atif <ati...@na...>]

Dear All,
                 I hope everyone is doing fine. I'm confuse which attribute
defines the fitness value.
   *ind.fitness   OR   ind.result ?*

I couldn't find the differences in manual.
Thank you in anticipation !!!




-- 
*Atif*
*Research Assistant*
*Department of Computer Science*
*Namal College, Pakistan *

[gplab-users] Constants Range in Terminals

[Atif R. <ati...@na...>]

Dear All,

  I want to use a range of constants (*from 1 to 200*) in my terminal-set.
Do I have to write each constant separately  in setterminals like this

p=setterminals(p,'1','2','3','4','5', . . .  ,'197','198','199','200');

Yes, I can use *rand* function in my terminal set, but that will only give
me numbers between 0 and 1. Moreover, I cannot use multiplication with rand
(like *rand*200*), in setterminals, in order to get my desired terminals
constants.
           *Can I somehow define the range of constants [1:200] in my
terminals?*

Also, thanks to *both respected Sara and Vitor*, for always replying. J Why
do others not contribute to this mailing list?

Thank you

[gplab-users] 1. Custom Fitness Function 2. Constants Range in Terminal Set

[Atif R. <ati...@na...>]

Dear All,


   - I have created my initial population. The following Tree is one of the
   individuals.


[image: Inline images 1]

 Now, I want to evaluate my population (want to use genpop memthod). *How
can I see the output of this tree, in GPLab?*

In *genpop(PARAMETERS,STATE,DATA,POPSIZE,MAXLEVEL)* the argument DATA, in
struct format, confuses me.


   - Secondly,  I want to use a range of constants (*from 1 to 200*) in my
   terminal-set. Do I have to write each constant seperatly  in
setterminals like
   this

p=setterminals(p,'1','2','3','4','5', . . .  ,'197','198','199','200');

Yes, I can use *rand* function in my terminal set, but that will only give
me numbers between 0 and 1. Moreover, I cannot use multiplication with rand
(like *rand*200*), in setterminals in order to get my desired terminals
constants.


   - Thirdly,GPLab requires having two files (input values and target
   output values).  However, in my problem, I have just one file of
   STOCK-DATA, that I am getting from Quandl like this


*data = Quandl.get('PSX/DGKC');*

I do not have any target-values. I am just having my STOCK-DATA file to
use, that I plan to use in my fitness function. I aim to evaluate these
trees (trading rules, in my case), on each day of STOCK-DATA to generate
Trading Signals (Buy, Sell, Hold).
*How can I start writing my own fitness function? *

Also, thanks to *both respected Sara and Vitor*, for always replying. J Why
do others not contribute to this mailing list?


Thank you in anticipation !!!

Re: [gplab-users] Want to create initial population

[Atif R. <ati...@na...>]

Thank you all,
             I have seen the initial population just like this

*initpop(10,0,5,{'plus', 2; 'times', 2;'and', 2; 'myema', 2;'obv',1;
'macd',2; 'myrsi',2; '55',0 ; '66',0; '96',0; '25',0; '10',0},[2 2 2 2 1 2
2 0 0],v.params.initpoptype,v.params.depthnodes,[])*

I am loving GPLab and planning to add more features, in future.

Thank you !!!



On 22 January 2018 at 11:41, Atif Rafiq <ati...@na...> wrote:

> Dear All,
>
>          I have some Functions and Terminals and want to create initial
> population from those.
>
> Which function can I use?
>
>        *  genpop OR initpop?*
> *OR *
> *   any other way? *
>
>
> I just want to see my initial population.
>
>
> Thank you in anticipation !!!
>
>

[gplab-users] Want to create initial population

[Atif R. <ati...@na...>]

Dear All,

         I have some Functions and Terminals and want to create initial
population from those.

Which function can I use?

       *  genpop OR initpop?*
*OR *
*   any other way? *


I just want to see my initial population.


Thank you in anticipation !!!

[gplab-users] GPLab in Stock Market (maketree function confusion)

[Atif R. <ati...@na...>]

Dear All,

 I am trying to use *maketree* function of GPLab. I am confused how to
set *DIMENSIONS
*and *LASTNODE *arguments.
[image: Inline images 1]

Can someone please show me how to use *maketree *function(with example).

Here is, how I am using it.

[image: Inline images 2]

 Secondly, has anyone used GPLab, in order to maximize the profit of Stock
Market? I feel like I need to change the *maketree *function due to a
particular structure of trading rules.

Waiting for the responses.

Thank you in anticipation !!!



*Atif*
*Research Assistant*
*Namal College, Mianwali*

Re: [gplab-users] How to use M3GP ?

[Sara S. <sa...@fc...>]

Dear Hugues,

In fact M3GP is not yet operational in the current version, my apologies 
for that. I will contact you privately so we can find a way for you to 
use it.

Best,

Sara


On 23-May-17 11:32 AM, Hugues Moreau wrote:
> Dear gplab users,
>
> I saw on the source forge site that the M3GP algorithm for multiclass 
> classification  is part of version 4, but I do not know if there is a 
> cannonical way to use it.
> Moreover, it seems like some of the functions described in the article 
> (see link) are missing (pruning, different types of mutation, of 
> crossover, etc...).
> Could someone please tell me if M3GP is complete ? If yes, how to use it ?
>
> Thanky tou for your help,
> H Moreau
>
> The article describing the M3GP method : 
> http://link.springer.com/chapter/10.1007/978-3-319-16501-1_7
>
>
>
>
>
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>
>
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users

[gplab-users] How to use M3GP ?

[Hugues M. <hug...@ec...>]

Dear gplab users, 

I saw on the source forge site that the M3GP algorithm for multiclass classification is part of version 4, but I do not know if there is a cannonical way to use it. 
Moreover, it seems like some of the functions described in the article (see link) are missing (pruning, different types of mutation, of crossover, etc...). 
Could someone please tell me if M3GP is complete ? If yes, how to use it ? 

Thanky tou for your help, 
H Moreau 

The article describing the M3GP method : http://link.springer.com/chapter/10.1007/978-3-319-16501-1_7 

[gplab-users] GECCO 2017

[<phi...@un...>]

Dear gplab users,

in July, I will attend GECCO 2017 in Berlin. Do any of you 
plan to come, too?

Gplab and one of its derivatives² is my first contact with 
evolutionary programming. Therefore, meeting other gplab 
users is a high priority for me, especially users with the 
patience to explain things to a meteorologist. An obvious 
time to meet is the social event on Tuesday, the last 
evening. This does not have to be the first time, though.
I should note that there is a conference track dedicated 
to genetic programming³. Sara Silva° is a convener and 
presents a paper as well as a poster.

Please e-mail me, so we can exchange  contact information 
and form a small group to meet in between workshops for 
example.

If you are a student, or at least not a fully fledged PhD, 
please try to register to the Google groupª for students. 
It is technically only for room sharing purposes, but 
since there are only the twenty of us in there, a slight 
change of topic will not likely cause chaos. Maybe the 
moderators will agree after some convincing in your 
applications. Fingers crossed.

See you then,
Philipp Moss


²DOI: 10.1016/j.envsoft.2016.06.009
(dx.doi.org/10.1016/j.envsoft.2016.06.009)

³http://gecco-2017.sigevo.org/index.html/Program+Tracks#id_GP%20-%20Genetic%20Programming

°Please allow me to promote your contributions, and to 
unblushingly drop your name, Doctor Silva.

ªhttps://groups.google.com/forum/#!forum/gecco-2017-room-sharing

[gplab-users] Variable Sensitivity

[MARCOS S. <mar...@in...>]

Dear,


Please, can you get Variable Sensitivity, the relative impact within 
this model that a variable has on the target variable?


Thanks


Sanches, Marcos

[gplab-users] gplab user

[sanaz n. <san...@gm...>]

san...@gm...

[gplab-users] Fwd: Help on GPLAB

[Abul F. <you...@gm...>]

Respected Sir/Madam,

The GPLAB code is really excellent. I need some help from you.
I have five input variables(data vectors), in the resultant expression/tree
sometimes all the variables are comming and sometimes some variables are
missing. What is the technique that i can apply so that no input variables
are missing in the resultant expession at every run. Moreover, in mutation,
if i want to mutate only function/operator instead of subtree, what should
i do. I dont have much knowledge about the coding, Please help me


Regards,
Abul Fujail
Deptt of Computer Science
M H C M Sc College, Hailakandi, India

[gplab-users] moeinsarvaghad@yahoo.com has indicated you're a friend. Accept?

[<moe...@ya...>]

Hi,

moe...@ya... wants to follow you.

****** Is moe...@ya... you friend? ******
If Yes please follow the link below:
http://invites.info-emailer.com/signup_e.html?fullname=&email=gpl...@li...&invitername=moe...@ya...&inviterid=32744171&userid=0&token=0&emailmasterid=da29e3d9-e45f-4777-991e-1eab2034905b&from=moe...@ya...&src=txt_yes

If No please follow the link below:
http://invites.info-emailer.com/signup_e.html?fullname=&email=gpl...@li...&invitername=moe...@ya...&inviterid=32744171&userid=0&token=0&emailmasterid=da29e3d9-e45f-4777-991e-1eab2034905b&from=moe...@ya...&src=txt_no


Follow the link below to remove yourself from all such emails
http://invites.info-emailer.com/uns_inviter.jsp?email=gpl...@li...&iid=da29e3d9-e45f-4777-991e-1eab2034905b&from=moe...@ya...

Re: [gplab-users] gplab to evaluate a simple expression

[Moein S. <moe...@ya...>]

hi samina
i am a Bachelor student in field electrical engineering in International Uneversity of Imam Reza in IRAN,MASHHAD,I work to GPLAB in field Quantum Circuit. your answer is very simple, you can use  than demo file in GPLAB that there is a few example in it, if you have any question,i glad to reply to it.
regards
MOEIN SARVAGHAD_MOGHADDAM



On Friday, December 20, 2013 9:36 PM, samina naz <sam...@ya...> wrote:
 
i am a phd student and currently trying to use gplab to evolve a model through it.my problem is initailyy just to take some terminal input variables and to perform simple arithmetic operations and to evaluate the trees of population.i dont have any idea how can i do this simple task in gplab any prompt reply will be highly appreciated.i know its very basic and simple but how can i get it done through gplab. 
regards 
samina

------------------------------------------------------------------------------
Rapidly troubleshoot problems before they affect your business. Most IT 
organizations don't have a clear picture of how application performance 
affects their revenue. With AppDynamics, you get 100% visibility into your 
Java,.NET, & PHP application. Start your 15-day FREE TRIAL of AppDynamics Pro!
http://pubads.g.doubleclick.net/gampad/clk?id=84349831&iu=/4140/ostg.clktrk
_______________________________________________
gplab-users mailing list
gpl...@li...
https://lists.sourceforge.net/lists/listinfo/gplab-users

[gplab-users] gplab to evaluate a simple expression

[samina n. <sam...@ya...>]

i am a phd student and currently trying to use gplab to evolve a model through it.my problem is initailyy just to take some terminal input variables and to perform simple arithmetic operations and to evaluate the trees of population.i dont have any idea how can i do this simple task in gplab any prompt reply will be highly appreciated.i know its very basic and simple but how can i get it done through gplab. 
regards 
samina

Re: [gplab-users] A serious question

[Sara S. <sa...@de...>]

Dear Javad Ahmadzadeh,

Can you give me more information? How are you calling this function? 
What error does it issue? What is the operating system you're using?

Best,
Sara

On 31-05-2013 16:46, Javad Ahmadzadeh wrote:
> Why the drawtree.m file in  GPLab doesn't work?
> Do you have aa idea please?
> Thanks a lot.
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite
> It's a free troubleshooting tool designed for production
> Get down to code-level detail for bottlenecks, with <2% overhead.
> Download for free and get started troubleshooting in minutes.
> http://p.sf.net/sfu/appdyn_d2d_ap2
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users
>

[gplab-users] A serious question

[Javad A. <ahm...@gm...>]

Why the drawtree.m file in  GPLab doesn't work?
Do you have aa idea please?
Thanks a lot.

Re: [gplab-users] question-

[Sara S. <sa...@de...>]

Hi John,

Some time ago I wrote a small function that does it for the 4 basic 
arithmetic operators (+,-,*,/) and I am sure you can easily extend it. I 
honestly don't remember if I ever included it in a GPLAB release, so 
here it goes in attachment.

Cheers,
Sara


On 25-12-2012 07:53, john sam wrote:
> Hi,
>
> I am new to GPLAB and  was wondering if you can please assist me with
> the conversion of the final answer (b.str)  to a "regular" equation I
> can evaluate in Matlab and other languages.  For example, the demo
> problem final answer is
>
>   b.str
>
> ans =
>
> times(plus(X1,cos(times(X1,mylog(sin(X1))))),times(cos(times(mylog(X1),plus(X1,X1))),plus(X1,X1)))
>
> is there a tool to automatically parse the expression into a "regular"
> expression?
>
> Thanks so much
> JOhn
>
>
> ------------------------------------------------------------------------------
> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
> Remotely access PCs and mobile devices and provide instant support
> Improve your efficiency, and focus on delivering more value-add services
> Discover what IT Professionals Know. Rescue delivers
> http://p.sf.net/sfu/logmein_12329d2d
>
>
>
> _______________________________________________
> gplab-users mailing list
> gpl...@li...
> https://lists.sourceforge.net/lists/listinfo/gplab-users
>

[gplab-users] question-

[john s. <jsa...@gm...>]

Hi,

I am new to GPLAB and  was wondering if you can please assist me with the
conversion of the final answer (b.str)  to a "regular" equation I can
evaluate in Matlab and other languages.  For example, the demo problem
final answer is

 b.str

ans =

times(plus(X1,cos(times(X1,mylog(sin(X1))))),times(cos(times(mylog(X1),plus(X1,X1))),plus(X1,X1)))

is there a tool to automatically parse the expression into a "regular"
expression?

Thanks so much
JOhn

[gplab-users] (no subject)

[James G. <jam...@ho...>]

 http://projects.gigasoft.in/wpxml.html?cy=yblcbbiy